| ▲ | roadbuster a day ago | ||||||||||||||||||||||||||||||||||||||||
"Processor-in-Memory", is an age-old idea which was even explored in a doctoral thesis from 1998: "Computational RAM: A Memory-SIMD Hybrid"*. At the time, the motivation was, "Between a DRAM chip and a microprocessor mounted on a circuit board, we can only get a few hundred megabytes per second of bandwidth. However, inside the memory chip itself, there's hundreds of gigabytes (terabits) of bandwidth available, so why not move some of the logic there?". And then the author pushed it to the extreme where there was a wide array of 1-bit processors at each column of DRAM. (Note: the reason why true PIM isn't viable is because of the radically different semiconductor process used for DRAM vs. high-speed digital logic, i.e.: your digital logic circuits will perform poorly on a "DRAM process"). The designers of HBM5 had the same observations and thought, "why don't we put a logic chip at the bottom of the stack of HBM DRAM chips and get extremely high memory bandwidth?" But, in looking at this, you'll also quickly realize this is almost exactly what Intel and Micron pitched 15 years ago as Hybrid Memory Cube** (which lost on the market compared to the simpler, less costly, and more flexible alternative, HBM) * https://www.eecg.toronto.edu/~stumm/Theses/Elliott-PhD98.pdf | |||||||||||||||||||||||||||||||||||||||||
| ▲ | snailmailman 18 hours ago | parent | next [-] | ||||||||||||||||||||||||||||||||||||||||
I think it goes back further than 1998. Several old systems had special memory locations where the numbers written would then read back changed. The Apollo Guidance computer for instance didn’t have a shift-right or shift-left cpu instruction. They would write data to a specific area of memory that would modify the data to shift bits around. Although I’m sure it’s quite different. The cpu and memory hardware is so completely different. And basic bit manipulation is hardly computation. | |||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||
| ▲ | hn_acc1 a day ago | parent | prev | next [-] | ||||||||||||||||||||||||||||||||||||||||
Duncan Elliott taught my graduate (IIRC) computer architecture course while writing that thesis - we all talked about C-RAM for a while.. And I think it was a rehash of an even older paper / discussion about that idea. | |||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||
| ▲ | timcobb a day ago | parent | prev [-] | ||||||||||||||||||||||||||||||||||||||||
Age-old... 1998... :grimace: | |||||||||||||||||||||||||||||||||||||||||
| |||||||||||||||||||||||||||||||||||||||||